A Novel Frequency Characteristic Model and Noise Shaping Method for Encoder-Based Speed Measurement in Motor Drive

The performance of the closed-loop control system is determined by the accuracy of the feedback signal to some extent. In most motor drive systems, the position and speed signals come from the optical encoder. Thus, to reduce the quantization error from encoder and achieve better speed measurement results are the focuses of scholars and engineers. Although many improved speed measurement methods have been proposed, there are still few literatures discussed about the frequency characteristic of the encoder quantization noise. Even in some algorithms like Kalman filter based speed measurement methods usually regard the quantization noise as white noise. In this paper, a noise frequency characteristic model of encoder-based speed measurement in motor drive is built up. According to this model, the variation rule of the noise frequency is found. Besides, a variable-period noise shaping method is proposed to reduce the impact of quantization noise. Both the notch filter and a finite impulse response (FIR) filter are used to cope with the quantization noise. The frequency model and noise shaping method are verified by experiment results.